tradermonty/claude-trading-skills

What it is

Claude Trading Skills packages a set of Claude Skills aimed at individual investors who run a long-term core (ETFs, dividend stocks) and treat swing trading as a satellite activity. The stated goal is process, not prediction: the skills structure market-regime review, portfolio checks, screening, position sizing, pre-trade discipline, journaling, and post-trade review rather than issuing buy/sell signals. The repository is explicit that it is not a signal service, broker execution platform, or automated trading system, and that human decision gates remain central.

The skills are grouped by area — market regime, core portfolio, dividend income, swing opportunities, trade planning, trade memory, strategy research, and advanced satellite tools — with each skill living in its own folder containing a SKILL.md, reference material, and helper scripts. skills-index.yaml is the canonical metadata index, and multi-skill routines are defined as manifests under workflows/ workflow manifests. A Trading Skills Navigator skill recommends a starting path from a plain-language goal navigator.

Quick start

Two installation routes are documented. For the Claude web app: download the matching .skill archive from the latest release, verify it against SHA256SUMS, enable Code execution and file creation under Settings > Capabilities, then upload it under Customize > Skills web app steps. For Claude Code, clone the repository and copy a skill folder (for example backtest-expert) into ~/.claude/skills/ for personal use or .claude/skills/ inside a project Claude Code setup.

A documented no-paid-API path uses five skills manually: market-breadth-analyzer, uptrend-analyzer, position-sizer, trader-memory-core, and signal-postmortem. This still needs public CSVs, chart screenshots, or local files as inputs; check each skill's integrations: entry in the skills index for exact requirements. A guided walkthrough from installation to first journal entry is in Your First Week.

Requirements and constraints

The shared project environment supports Python 3.11–3.13 on Linux, Windows, and macOS, while packaged standalone skills retain a Python 3.9 source floor; only combinations exercised by CI count as supported, with the rest best-effort as described in the compatibility matrix. Claude Code has separate account requirements from Claude.ai, and the Free plan does not include Claude Code. FMP, FINVIZ Elite, and Alpaca are optional paid integrations for specific skills: FMP is needed by many screeners, FINVIZ Elite costs a monthly or annual subscription, and Alpaca (with its MCP server) is used for portfolio holdings. The project is MIT-licensed and provided AS IS, and it states repeatedly that it is for educational and research use, not financial advice.

HotRepo's take

The repository reached 9th place on the all-language GitHub daily ranking on 2026-11-13, adding 60 stars that day per the snapshot. What stands out is the delivery format: these are Claude Skills with a canonical YAML index, workflow manifests, schema validation, and sample runs, not a monolithic application. That structure makes the project's usefulness depend on how many skills a reader actually installs and feeds with data, so the no-API five-skill path is the honest entry point for evaluating it before adding paid data subscriptions.